Accumulating evidence indicates that there is great variability in the use of medical resources, perhaps particularly in ambulatory care. The evidence also indicates that this elasticity cannot be attributed primarily to clinical differences among patients, but rather to the variable "styles" of differet practitioners in responding to "clinically-unrelated" pressures on their time, effort and income. While there has been an enormous investment made in understanding the pathophysiologic basis of disease and its treatment, relatively little attention has been paid to the process by which clinical decisions are made. Yet these decisions by individual practitioners have a profound impact on both the quality and cost of care. We believe that it is possible to enhance clinical decision making, and thus the quality and cost of care, through the design and use of cost-effective strategies for common clinical conditions. In this proposal, we describe an effort to design such strategies for several of the most common conditions seen in ambulatory care setting. The design of these strategies would employ an intensive search of the published medical literature, the use of information previously collected by our group and others in computerized clinical databases, the application of quantitative approaches to decision making (e.g., probability theory, multivariate analysis of discrete variables, and decision analysis), and the solicitation of opinion from various national experts when published empirical data is absent.